From 763ea86c98e3dcc2f797011514d21d4c257ce2ab Mon Sep 17 00:00:00 2001
From: Simon Meers <simon@simonmeers.com>
Date: Mon, 15 Jul 2024 14:56:34 +1000
Subject: [PATCH] update Saskatchewan

---
 ca_sk/people.py | 113 +++++++++++++++++++++++-------------------------
 1 file changed, 54 insertions(+), 59 deletions(-)

diff --git a/ca_sk/people.py b/ca_sk/people.py
index 6f46fe52..b6a584b7 100644
--- a/ca_sk/people.py
+++ b/ca_sk/people.py
@@ -1,3 +1,5 @@
+import re
+
 from utils import CanadianPerson as Person
 from utils import CanadianScraper
 
@@ -8,71 +10,64 @@ class SaskatchewanPersonScraper(CanadianScraper):
     def scrape(self):
         page = self.lxmlize(COUNCIL_PAGE)
 
-        members = page.xpath('//table[@id="MLAs"]//tr')[1:]
+        members = page.xpath('//table[@id="mla-table"]//tr')[1:]
         assert len(members), "No members found"
         for member in members:
-            if "Vacant" not in member.xpath("./td")[0].text_content():
-                name = member.xpath("./td")[0].text_content().split(". ", 1)[1]
-                district = member.xpath("./td")[2].text_content()
-                url = member.xpath("./td[1]/a/@href")[0]
-                page = self.lxmlize(url)
-                party = page.xpath('//span[@id="ContentContainer_MainContent_ContentBottom_Property4"]' "/span")[
-                    0
-                ].text
-
-                p = Person(primary_org="legislature", name=name, district=district, role="MLA", party=party)
-                p.add_source(COUNCIL_PAGE)
-                p.add_source(url)
-                try:
-                    p.image = page.xpath('//div[contains(@class, "mla-image-cell")]/img/@src')[0]
-                except IndexError:
-                    pass
+            if "Vacant" in member.xpath("./td")[1].text_content():
+                continue
+            name = member.xpath("./td")[0].text_content().split(". ", 1)[1].strip()
+            district = member.xpath("./td")[2].text_content().strip()
+            url = member.xpath("./td[1]/a/@href")[0]
+            page = self.lxmlize(url)
+            party = page.xpath('//div[contains(@class, "mla-header")]')[0].text.split(' - ')[1].strip()
 
-                contact = page.xpath('//div[@id="mla-contact"]/div[2]')[0]
-                website = contact.xpath("./div[3]/div[3]/div[2]/a")
-                if website:
-                    p.add_link(website[0].text_content())
+            p = Person(primary_org="legislature", name=name, district=district, role="MLA", party=party)
+            p.add_source(COUNCIL_PAGE)
+            p.add_source(url)
+            try:
+                p.image = page.xpath('//div[contains(@class, "mla-image-cell")]/img/@src')[0]
+            except IndexError:
+                pass
 
-                def handle_address(lines, address_type):
-                    address_lines = []
-                    for line in lines:
-                        if line.endswith(":"):  # Room:, Phone:, Fax:
-                            break
-                        address_lines.append(line)
-                    if address_lines:
-                        p.add_contact(
-                            "address",
-                            " ".join(address_lines),
-                            address_type,
-                        )
+            def handle_address(lines, address_type):
+                address_lines = []
+                for line in lines:
+                    if re.match(r'(Room|Phone|Fax)\:', line):
+                        break
+                    address_lines.append(line)
+                if address_lines:
+                    p.add_contact(
+                        "address",
+                        " ".join(address_lines),
+                        address_type,
+                    )
 
-                def handle_phone(lines, phone_type):
-                    if "Phone:" in lines:
-                        next_line = lines[lines.index("Phone:") + 1]
-                        if next_line.endswith(":"):
-                            return
-                        number = None
-                        if "/" in next_line:
-                            for fragment in next_line.split("/"):
-                                if fragment.strip().startswith("306-"):
-                                    number = fragment.strip()
-                                    break
-                        else:
-                            number = next_line
-                        p.add_contact("voice", number, phone_type, area_code=306)
+            def handle_phone(lines, phone_type):
+                matches = re.findall(r'Phone\:\s*(306-[\d\-]+)', '\n'.join(lines))
+                if len(matches) == 1:
+                    p.add_contact("voice", matches[0], phone_type, area_code=306)
 
-                legislature_lines = contact.xpath('.//div[@class="col-md-4"][1]/div//text()')
-                assert legislature_lines[0] == "Legislative Building Address"
-                handle_address(legislature_lines[1:], "legislature")
-                handle_phone(legislature_lines[1:], "legislature")
+            for address in page.xpath('//div[@class="col-md-3"]'):
+                lines = address.xpath('./div//text()')
+                address_type = None
+                if lines[0] == "Legislative Building Address":
+                    address_type = "legislature"
+                elif lines[0] == "Constituency Address":
+                    address_type = "constituency"
+                else:
+                    raise AssertionError(f"Unexpected address type: {lines[0]}")
+                handle_address(lines[1:], address_type)
+                handle_phone(lines[1:], address_type)
 
-                constituency_lines = contact.xpath('.//div[@class="col-md-4"][2]/div//text()')
-                assert constituency_lines[0] == "Constituency Address"
-                handle_address(constituency_lines[1:], "constituency")
-                handle_phone(constituency_lines[1:], "constituency")
+            email = self.get_email(page.xpath('//div[@id="content"]')[0], error=False)
+            if email:
+                p.add_contact("email", email)
 
-                email = self.get_email(contact, error=False)
-                if email:
-                    p.add_contact("email", email)
+            websites = re.findall(
+                r'Website:\s*(http\S+)',
+                ' '.join(page.xpath('//div[@class="col-md-4"]/div//text()'))
+            )
+            if len(websites) == 1:
+                p.add_link(websites[0])
 
-                yield p
+            yield p